Electro-optic Mach-Zehnder modulators are ubiquitous devices found in high speed optical networking equipment and are the key components that allow the transmission of high speed data using a beam of light. One such type of electro-optic modulator (EOM) uses lithium niobate (LiNbO3) crystals due to their low optical loss characteristics and high electro-optic coefficient. These modulators have a characteristic transmission response that is a function of the applied voltage, and typically require a direct current (DC) bias voltage to maintain the transmission response at a point that yields optimum transmission characteristics. However, one of the problems with lithium niobate modulators is that the transmission response, and thus the bias voltage, drifts over the long term due to temperature dependence and aging, requiring the use of a closed loop feedback circuit to maintain the optimum operating bias point. Conventional systems and methods for maintaining the optimum operating bias point utilize a low frequency dither signal imposed on the high frequency data signal, which is recovered in a feedback circuit. This is typically accomplished using complex analog circuitry. These systems and methods typically incorporate a low frequency sine, square, or triangular wave to amplitude modulate the high frequency RF data signal driving the modulator. An optical tap from the modulated output is passed through a transimpedance amplifier to convert the optical output signal to a voltage. A bandpass filter is used to recover either the fundamental tone or a harmonic of the original square wave. If the fundamental tone is recovered, the signal is synchronously detected and passed through a full wave rectifier. The resulting signal is then passed through an integrator which, ideally, has infinite DC gain and the resulting error signal is used to create the DC bias voltage for the modulator.
Thus, the conventional systems and methods for maintaining the optimum operating bias point utilize complex analog circuitry, a dither tone generation circuit, a transimpedance amplifier, a bandpass filter, and some sort of synchronous detection of the recovered signal. An error amplifier/integrator with low loop bandwidth is used to control the operating bias point.
There are several important drawbacks that are associated with the conventional systems and methods for maintaining the optimum operating bias point that are addressed by the systems and methods of the present invention. First, the conventional systems and methods require the use of multiple electronic components that take up board space and increase cost. Second, the conventional systems and methods require the synchronous detection of the recovered dither tone. This is typically accomplished by using the original dither signal to sample the recovered dither tone. Because the recovered dither tone is passed through a bandpass filter, there is a phase difference between the original dither signal and the recovered dither tone. In order to synchronize the detection, a phase adjustment of the sampling signal may be required to begin the sampling precisely at the start of each period of the recovered dither tone. This increases the complexity of the circuit. Third, the conventional systems and methods require that a bandpass filter be used to recover the dither tone. In order for this bandpass filter to be physically realizable, the dither tone must be of sufficiently high frequency such that the filter components are not too large; however, the frequency must not be so high such that it perturbs the data signal. Similarly, the dither tone must be of sufficiently high amplitude such that it may be recovered, however, the amplitude must not be so high such that it perturbs the data signal. Fourth, the conventional systems and methods require the use of a reset switch, which is typically software-controlled, to reset the integrator to avoid saturation. This may cause the modulator to become “stuck” at either extreme value of the integrator.
Thus, what are needed are improved software-based electro-optic modulator bias control systems and methods that utilize optical power slope detection and RF drive level optimization.